By
Siri Mehus, Ph.D.
At Blink UX we perform qualitative research to inform the design of digital products. Focus groups are a familiar form of qualitative consumer research. But while the focus group has a place in our research methods toolkit, we rarely pull it out. Why is that?
There are two main ways in which focus groups differ from the user research types we usually conduct, such as usability testing or contextual interviews.
- A focus group includes multiple participants (usually around 8-15), rather than one individual.
- Focus groups investigate what people believe, feel, or perceive, not what they do or why they do it.
While there may be other differences between focus groups and primary UX research methods, these two dimensions are enough to understand why a focus group is often not the right approach for product design research.
Saying vs. doing
In a focus group participants who do not know each other gather in a conference room or similar space to answer questions and discuss their thoughts about a topic, such as a consumer product. Due to the limitations of the group setting, their opportunity to interact with the product is severely constrained. Further, the discussion occurs in a social and physical environment that is completely unrelated to the one in which they would ordinarily encounter the product. As such, their reactions to the product are hypothetical and abstracted from real-life use.
Because focus group participants are not in a position to actually show us what they would do, our only source of information is what they tell us. And as Jakob Nielsen and Blink’s own Brian Essex remind us, when it comes to product design, listening to what people say is far less useful than watching what people do.
Why? Because people are not good at predicting what they’ll do in the future, remembering what they’ve done in the past, or explaining why they do what they do. Most are also not good at designing products—that’s why designers train for years. So asking them questions about what they usually do, what they would do, or how a product should look or work will yield results of limited value.
Here’s what makes this really dangerous: Although people don’t really know what they’ve done, what they will do, or how to design a product, that won’t stop them from giving you answers if you ask them those questions. So if you use this technique you will generate findings that superficially appear to be illuminating, but could potentially lead your team in the wrong direction.
On the other hand, people are really good at behaving like people, especially in realistic situations. That’s why the most effective research methods involve observing people interacting with products in natural environments (field studies) or performing representative tasks with prototypes or products (in-lab usability testing). These methods allow us to see what users actually do, rather than depending on unreliable self-reports of what they have done or would do.
Many vs. one
A seeming advantage of the focus group is the ability to get feedback from many people quickly and at less expense than if you were to talk to participants one at a time.
However, it is important to note that there are big differences between a one-to-one conversation and a group interaction.
Interacting in a group is fundamentally different from interacting in a focused dialogue between two people. That difference will have an effect on who speaks, how they speak, and—most importantly—what they speak about.
In a focus group, what people say is influenced by the presence of the other participants. For instance, participants may be hesitant to reveal confusion or ignorance. More talkative group members can dominate the conversation and prevent the more reticent from being heard. Participants’ opinions can be influenced by what they hear from other group members. And there is insufficient time for a moderator to probe and ask clarifying questions to fully understand the positions of individual members.
Blink research director Tom Satwicz provides the following example of what could be missed by a focus group asked to assess whether a design is “on the right track.”
Let’s say the group was shown mock-ups of an app that included a “hamburger” icon. Towards the beginning of the session, one participant noticed that feature and commented that he was familiar with it and expected to find menu items there. After that, the conversation moved on to other topics and the hamburger was not mentioned again.
How many of those participants would have known what that icon meant and how to use it? Would that icon have presented a significant obstacle to users’ ability to accomplish their goals? Was the design, in fact, on the right track? Those questions would not have been answered by the focus group.
All of this means that the feedback received from participants in a focus group cannot be taken as representative of how individuals would react on their own. As such, results should not be given any more weight than that gained from a single usability session, and perhaps less.
When are focus groups a good idea?
Focus groups have a long history of use for research on branding and marketing campaigns, and they may well be the right option for that purpose. Even for product design research there are situations in which a focus group could be an appropriate choice.
Because it is not possible to actually observe people using a product during a focus group, they are better employed for learning about thoughts and perceptions. This information is most useful very early in the process, before anything is designed or built. A focus group might help you gather insights into the aspects of an idea that people find exciting or possibly the aspects they find uninteresting or objectionable. However, because of the small sample and biasing effects of the group interaction format, the findings that emerge from a focus group should not be taken as representative of what people in general (or a target user group) want or like, any more than would an interview with a single participant.
The group format allows for participants to not only describe their thoughts and opinions to a researcher, but also to discuss and even debate them among themselves. This allows you to see participants’ reactions to ideas they may not have even considered if they were on their own. While this moves us further away from an authentic understanding of how users will react to a product, it may generate a greater number of reactions and ideas. In this sense a focus group can be much like a brainstorming session: if you seek to maximize the number of ideas proposed and are not concerned that they all be high quality (i.e., representative of what people would actually do or think in a realistic situation), a focus group may be a good choice.
Finally, a focus group can be a relatively inexpensive option when the alternative is no research at all. However, low cost should not be assumed. Like usability testing, running a focus group requires recruiting time, a trained moderator, and an appropriate space. Talking to ten people in a single group may be no less expensive than talking to two or three people individually.
The Dyad Alternative
An intermediate solution is to interview participants in pairs, rather than individually or in groups. This doubles the overall number of participants interviewed, but will decrease the time spent with each individual. As with a focus group, it allows individuals to consider ideas that they might not have come up with on their own. While it is still possible for participants to influence one another, a skilled moderator will be more able to tease out individual perspectives by probing and asking follow up questions of each participant.
For the most part, this type of interview would be used to gather participant thoughts and opinions, rather than to observe participants interacting with an actual product. However, viewing a couple of people perform a task together can be extremely illuminating when that is a reasonable use case (for instance, unboxing and setting up a new home technology product). Through their interaction with one another in the context of such an activity, participants verbally display their understandings of the task at hand and the obstacles they are encountering – providing the observing researcher with the natural version of a thinkaloud protocol.
Thoughts about how and when to effectively use focus groups? Questions about what research methods will deliver the best insights for your purposes? We’d love to hear from you! Contact us today!
Siri is a Principal UX Researcher at Blink with a true passion for finding the right methods to answer our clients’ product design questions. The best thing about playing matchmaker with methods? If the perfect fit doesn’t yet exist, we can invent it!